Gene Therapy for Chronic HBV—Can We Eliminate cccDNA?

Bloom, Kristie; Maepa, Mohube Betty; Ely, Abdullah; Arbuthnot, Patrick

Use of RGNs is now the most popular method of inactivating HBV gene expression. This bacterial CRISPR/Cas9 system relies on an RNA guided DNA binding domain with associated Cas9 endonuclease [48] . These RGNs typically comprise a CRISPR RNA (crRNA) with sequences that are complementary to a pre-defined DNA target sequence and a trans-activating crRNA (tracrRNA). Annealing of the RNA to its cognate enables Cas9-mediated double-stranded target DNA cleavage. Fusing crRNA and tracrRNA to form a single guide RNA (gRNA) has further simplified the system, which has made RGNs the easiest nucleases to engineer. Since first reported by Lin et al. in 2014 [46] , more than 16 publications have described anti-HBV efficacy of RGNs [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] (Table 1) . Targeted mutagenesis of cccDNA in mammalian cell cultures was first demonstrated by Seeger and Sohn [49] , who designed CRISPR/Cas9 constructs to bind the conserved enhancer II/core promoter and precore ORFs. Subsequent studies confirmed that single or multiple gRNAs spanning the entire HBV genome cleaved cccDNA to cause indel formation and resultant reduced viral protein expression [50] [51] [52] [54] [55] [56] 59, [62] [63] [64] [65] . However, methods used to detect mutagenic events vary greatly between individual studies, which makes it difficult to compare efficiency of different gRNA constructs directly. Nevertheless, as with all designer nucleases, targeting conserved regions is advantageous as antiviral efficacy is achieved across different genotypes [53, 55] . A second study by Seeger and Sohn used next generation sequencing to map CRISPR/Cas9-induced cccDNA mutations [58] . The majority of indels were identified as small deletions when extremely high targeted cleavage (up to 90%) was achieved. Although off-target cleavage may be frequently associated with CRISPR/Cas9 action [66] [67] [68] , data on the potential of HBV RGNs to create nonspecific mutations is scarce. Using multiple gRNAs to improve antiviral efficacy [46] or to excise integrated viral DNA fragments ( Figure 1A) [62] may exacerbate this issue. As expected off-target cleavage events were detected by deep sequencing when multiple gRNAs were used [57, 62] . However, other studies report no detectable genotoxic activity [54, 61, 63] , which may be a result of differences between gRNA design and properties of the viral target sequences. Replacing the Cas9 endonuclease with engineered nickases may improve specificity of HBV RGNs. With this configuration, gRNA and Cas9 nickases comprising heterodimers are required to effect double stranded DNA cleavage [51, 57] . Another challenge with the CRISPR/Cas9 system relates to delivery of the combined gRNA and the large Cas9 endonuclease. Sequences encoding these two components exceed the transgene capacity of the popular AAVs. A recent publication by Scott et al. demonstrated that the smaller Cas9 derived from Staphylococcus aureus (Sa), together with expression cassettes encoding short gRNAs targeting the HBV surface ORF, could be packaged into single stranded AAVs (ssAAV) [63] . Efficient delivery and expression of the RGN-encoding ssAAVs in HepG2.2.15 and hNTCP-HepG2 cells resulted in decreased viral replication, target specific cccDNA cleavage, and reduced cccDNA copy numbers. Analysis of sequences at predicted off target sites did not reveal non-specific mutagenesis.